On Monday 09 July 2007 13:04, Helena Mitasova wrote: > On Jul 9, 2007, at 3:47 PM, Dylan Beaudette wrote: > > On Wednesday 04 July 2007 08:41, Glynn Clements wrote: > >> "Carlos \"Guâno\" Grohmann" wrote: > >>> Interesting. > >>> Why r.slope.aspect uses > >>> > >>> dx = ... / (4 * ewres); > >>> > >>> while r.shaded.relief uses > >>> > >>> dx = ... / (8 * ewres); > >>> > >>> ?? > >> > >> Oops. It should be 8; the ewres/nsres values are actually the > >> distances between the centres of the top/bottom rows and left/right > >> columns in the 3x3 window, which are two rows/columns apart. IOW, the > >> 4*ewres should have been 4*(2*ewres) = 8*ewres, ditto for nsres. > > > > Yikes, does this mean that previous calculations are now wrong? > > no - it is correctly implemented - you would have found that it is > wrong by now. > It is just that the variable that you divide with in the r.slope.aspect > uses 2*ewres (twice the resolution, not just one as is usual in > the published equations) - I got caught on that many times when checking > the r.slope.aspect code. > > Helena
Thanks for the quick reply! That is a relief (! no pun intended there). Cheers, Dylan _______________________________________________ grass-dev mailing list [email protected] http://grass.itc.it/mailman/listinfo/grass-dev
